For healthcare professionals
RACE scale (Rapid Arterial oCclusion Evaluation)
The RACE scale is a prehospital stroke score designed to flag a possible large vessel occlusion (LVO). It adds five findings, facial palsy, arm motor function, leg motor function, head and gaze deviation, and aphasia or agnosia, for a total of 0 to 9. A score of 5 or more is the published cut-off. This calculator is written for clinicians and EMS. It is not a self-check and not a diagnosis.
Score the five items
Items, points and level names as printed in Rafiemanesh and colleagues (Rafiemanesh 2023). The short test notes are our own. Nothing you enter leaves this page.
What you will see
The RACE total from 0 to 9 and whether it reaches the published cut-off of 5 for a possible large vessel occlusion.
Educational only, not medical advice. A score supports, and does not replace, clinical judgment and local stroke triage protocols.
Source
Perez de la Ossa N et al., Stroke 2014 (the original scale), cited only. Items, points and level names as printed in Rafiemanesh H et al., BMC Emerg Med 2023, and the cut-off of 5 as printed in Schlemm L and Schlemm E, BMC Neurol 2018, and Yildiz HY et al., Front Neurol 2026, all open access (CC BY 4.0).
Validated in
Developed for prehospital use in Catalonia, Spain (Perez de la Ossa 2014). In the open studies cited here: 805 patients brought by EMS with suspected stroke to emergency departments in Iran (Rafiemanesh 2023), 303 stroke code activations in an emergency department in Istanbul, Turkey (Yildiz 2026), and 3,021 adults treated for an anterior circulation occlusion in the Netherlands, with RACE rebuilt from admission NIHSS scores (Duvekot 2021).
Limits
A triage aid for clinicians and EMS, not a diagnosis. At a cut-off of 5 it misses a large share of occlusions, more so for distal (M2) occlusions, and many people at or above it have no occlusion. It does not tell a stroke from a bleed or a stroke mimic. Vessel imaging decides.
How the scale is scored
The RACE scale has five items: facial palsy (0 to 2), arm motor function (0 to 2), leg motor function (0 to 2), gaze (0 to 1) and aphasia or agnosia (0 to 2), for a total of 0 to 9 (Rafiemanesh 2023). Higher scores mean more severe deficits (Schlemm 2018). Perez de la Ossa and colleagues designed and validated the scale in 2014 to predict a large arterial occlusion before hospital (Perez de la Ossa 2014).
| Item | 0 | 1 | 2 |
|---|---|---|---|
| Facial palsy | Absent | Mild | Moderate to severe |
| Arm motor function | Normal to mild | Moderate | Severe |
| Leg motor function | Normal to mild | Moderate | Severe |
| Head and gaze deviation | Absent | Present | Not scored |
| Aphasia or agnosia | Normal | Moderate | Severe |
| Total | 0 to 9. A score of 5 or more is the cut-off for a possible LVO. | ||
Item names, points and level names follow the Methods and Table 2 of Rafiemanesh and colleagues (Rafiemanesh 2023). Aphasia or agnosia is scored according to the side of the weakness (hemiparesis) (Schlemm 2018), so only one of the two is scored for each patient. For exact test steps, use the original paper and your local protocol (Perez de la Ossa 2014).
The cut-off of 5
Studies of the RACE scale use a fixed cut-off of 5 or more to flag a likely large vessel occlusion (Schlemm 2018). In an Istanbul emergency department, 56.5% of patients with an LVO scored 5 or more, compared with 11.1% of patients without one (Yildiz 2026). In a Dutch registry of patients with a treatable occlusion, a score of 5 or more had a sensitivity of 0.75 overall but 0.48 for more distal M2 occlusions (Duvekot 2021). In Iranian emergency departments, the same cut-off detected acute ischemic stroke of any kind with a sensitivity of 50.18% and a specificity of 92.18% (Rafiemanesh 2023). So a score under 5 does not rule out an occlusion, and local protocols, time from onset and imaging guide the final decision.
Citation check
The item names, points, level names and cut-off on this page were checked against the full text of the open access papers above in Europe PMC, where each carries a Creative Commons Attribution 4.0 (CC BY 4.0) license: the items and points against Rafiemanesh (2023), and the range, side rule and cut-off of 5 against Schlemm (2018) and Yildiz (2026). The original paper by Perez de la Ossa (2014) is cited, not quoted.
Related stroke tools
For a three-item motor score from 0 to 5 with a cut-off of 4, see the Los Angeles Motor Scale (LAMS). For the three-sign face, arm and speech screen used before hospital, see the Cincinnati Prehospital Stroke Scale. To score the full neurological deficit on arrival, use the NIH Stroke Scale (NIHSS) calculator. For a five-item score with eye deviation and denial or neglect and a cut-off of 4, see FAST-ED. Other bedside scores are listed under For healthcare professionals. For a yes or no screen that needs arm weakness plus visual disturbance, aphasia or neglect, see the VAN stroke screen.
Sources
- Rafiemanesh H, Barikro N, Karimi S, Sotoodehnia M, Jalali A, Baratloo A. The Rapid Arterial oCclusion Evaluation (RACE) scale accuracy for diagnosis of acute ischemic stroke in emergency department: a multicenter study. BMC Emerg Med. 2023;23:51. doi:10.1186/s12873-023-00825-7. PMID 37226097. Open access (CC BY 4.0). Methods and Table 2: RACE items, points and level labels, total 0 to 9; at 5 or more, sensitivity 50.18% and specificity 92.18% for acute ischemic stroke. pmc.ncbi.nlm.nih.gov/articles/PMC10210284/
- Schlemm L, Schlemm E. Clinical benefit of improved prehospital stroke scales to detect stroke patients with large vessel occlusions: results from a conditional probability model. BMC Neurol. 2018;18:16. doi:10.1186/s12883-018-1021-8. PMID 29427993. Open access (CC BY 4.0). Methods: RACE scores 0 to 9, aphasia or agnosia according to the side of hemiparesis, fixed cut-off of 5 or more. pmc.ncbi.nlm.nih.gov/articles/PMC5807751/
- Yildiz HY, Gulcan K, Kozba SE, Ozurun I, Demirkiran Z, Krespi Y. Clinical utility of the RACE score for differentiating stroke from stroke mimics in the emergency department. Front Neurol. 2026;17:1832156. doi:10.3389/fneur.2026.1832156. PMID 42206302. Open access (CC BY 4.0). RACE range 0 to 9; conventional cut-off of 5 or more: 56.5% of patients with LVO versus 11.1% without. pmc.ncbi.nlm.nih.gov/articles/PMC13202718/
- Duvekot MH, Venema E, Lingsma HF, et al. Sensitivity of prehospital stroke scales for different intracranial large vessel occlusion locations. Eur Stroke J. 2021;6(2):194-204. doi:10.1177/23969873211015861. PMID 34414295. Open access (CC BY 4.0). Table 2: RACE 5 or more, sensitivity 0.75 in the full cohort and 0.48 for M2 occlusions. pmc.ncbi.nlm.nih.gov/articles/PMC8370085/
- Perez de la Ossa N, Carrera D, Gorchs M, Querol M, Millan M, Gomis M, et al. Design and validation of a prehospital stroke scale to predict large arterial occlusion: the rapid arterial occlusion evaluation scale. Stroke. 2014;45(1):87-91. doi:10.1161/STROKEAHA.113.003071. PMID 24281224. Original scale (cited, not quoted). doi.org/10.1161/STROKEAHA.113.003071
- Centers for Disease Control and Prevention. Signs and Symptoms of Stroke. www.cdc.gov/stroke/signs-symptoms/index.html
- National Institute of Neurological Disorders and Stroke (NIH). Transient Ischemic Attack (TIA). www.ninds.nih.gov/health-information/disorders/transient-ischemic-attack
Written and maintained by the uncovermyrisk.com editorial team. We do not have a medical reviewer; every claim links to its primary source above. Last reviewed: 2026-11-04. Spotted an error? See About.